Lab6

Load Packages

library(tidyverse)

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.5
✔ forcats   1.0.0     ✔ stringr   1.5.1
✔ ggplot2   3.5.1     ✔ tibble    3.2.1
✔ lubridate 1.9.4     ✔ tidyr     1.3.1
✔ purrr     1.0.4     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(tidymodels)

── Attaching packages ────────────────────────────────────── tidymodels 1.3.0 ──
✔ broom        1.0.7     ✔ rsample      1.2.1
✔ dials        1.4.0     ✔ tune         1.3.0
✔ infer        1.0.7     ✔ workflows    1.2.0
✔ modeldata    1.4.0     ✔ workflowsets 1.1.0
✔ parsnip      1.3.1     ✔ yardstick    1.3.2
✔ recipes      1.1.1     
── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
✖ scales::discard() masks purrr::discard()
✖ dplyr::filter()   masks stats::filter()
✖ recipes::fixed()  masks stringr::fixed()
✖ dplyr::lag()      masks stats::lag()
✖ yardstick::spec() masks readr::spec()
✖ recipes::step()   masks stats::step()

library(readr)
library(powerjoin)
library(skimr)
library(visdat)
library(ggpubr)
library(ggcorrplot)
library(workflowsets)
library(recipes)

# List all .txt files in the data folder
data_files <- list.files("data", pattern = "\\.txt$", full.names = TRUE)

# Read each file using read_delim and store in a list
camels_list <- map(data_files, read_delim, delim = ";")

Rows: 671 Columns: 12
── Column specification ────────────────────────────────────────────────────────
Delimiter: ";"
chr (3): gauge_id, high_prec_timing, low_prec_timing
dbl (9): p_mean, pet_mean, p_seasonality, frac_snow, aridity, high_prec_freq...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 671 Columns: 8
── Column specification ────────────────────────────────────────────────────────
Delimiter: ";"
chr (3): gauge_id, geol_1st_class, geol_2nd_class
dbl (5): glim_1st_class_frac, glim_2nd_class_frac, carbonate_rocks_frac, geo...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 671 Columns: 14
── Column specification ────────────────────────────────────────────────────────
Delimiter: ";"
chr  (1): gauge_id
dbl (13): q_mean, runoff_ratio, slope_fdc, baseflow_index, stream_elas, q5, ...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 671 Columns: 12
── Column specification ────────────────────────────────────────────────────────
Delimiter: ";"
chr  (1): gauge_id
dbl (11): soil_depth_pelletier, soil_depth_statsgo, soil_porosity, soil_cond...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 671 Columns: 7
── Column specification ────────────────────────────────────────────────────────
Delimiter: ";"
chr (1): gauge_id
dbl (6): gauge_lat, gauge_lon, elev_mean, slope_mean, area_gages2, area_geos...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 671 Columns: 10
── Column specification ────────────────────────────────────────────────────────
Delimiter: ";"
chr (2): gauge_id, dom_land_cover
dbl (8): frac_forest, lai_max, lai_diff, gvf_max, gvf_diff, dom_land_cover_f...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

# Join all data frames using power_full_join by gauge_id
camels_data <- reduce(camels_list, power_full_join, by = "gauge_id")

# Look at structure of the combined data
glimpse(camels_data)

Rows: 671
Columns: 58
$ gauge_id             <chr> "01013500", "01022500", "01030500", "01031500", "…
$ p_mean               <dbl> 3.126679, 3.608126, 3.274405, 3.522957, 3.323146,…
$ pet_mean             <dbl> 1.971555, 2.119256, 2.043594, 2.071324, 2.090024,…
$ p_seasonality        <dbl> 0.187940259, -0.114529586, 0.047358189, 0.1040905…
$ frac_snow            <dbl> 0.3134404, 0.2452590, 0.2770184, 0.2918365, 0.280…
$ aridity              <dbl> 0.6305587, 0.5873564, 0.6241114, 0.5879503, 0.628…
$ high_prec_freq       <dbl> 12.95, 20.55, 17.15, 18.90, 20.10, 13.50, 17.50, …
$ high_prec_dur        <dbl> 1.348958, 1.205279, 1.207746, 1.148936, 1.165217,…
$ high_prec_timing     <chr> "son", "son", "son", "son", "son", "jja", "son", …
$ low_prec_freq        <dbl> 202.20, 233.65, 215.60, 227.35, 235.90, 193.50, 2…
$ low_prec_dur         <dbl> 3.427119, 3.662226, 3.514262, 3.473644, 3.691706,…
$ low_prec_timing      <chr> "mam", "jja", "djf", "djf", "djf", "mam", "mam", …
$ geol_1st_class       <chr> "Siliciclastic sedimentary rocks", "Acid plutonic…
$ glim_1st_class_frac  <dbl> 0.8159044, 0.5906582, 0.5733054, 0.4489279, 0.308…
$ geol_2nd_class       <chr> "Basic volcanic rocks", "Siliciclastic sedimentar…
$ glim_2nd_class_frac  <dbl> 0.17972945, 0.16461821, 0.28701001, 0.44386282, 0…
$ carbonate_rocks_frac <dbl> 0.000000000, 0.000000000, 0.052140094, 0.02625797…
$ geol_porostiy        <dbl> 0.1714, 0.0710, 0.1178, 0.0747, 0.0522, 0.0711, 0…
$ geol_permeability    <dbl> -14.7019, -14.2138, -14.4918, -14.8410, -14.4819,…
$ q_mean               <dbl> 1.699155, 2.173062, 1.820108, 2.030242, 2.182870,…
$ runoff_ratio         <dbl> 0.5434375, 0.6022689, 0.5558590, 0.5762893, 0.656…
$ slope_fdc            <dbl> 1.528219, 1.776280, 1.871110, 1.494019, 1.415939,…
$ baseflow_index       <dbl> 0.5852260, 0.5544784, 0.5084407, 0.4450905, 0.473…
$ stream_elas          <dbl> 1.8453242, 1.7027824, 1.3775052, 1.6486930, 1.510…
$ q5                   <dbl> 0.24110613, 0.20473436, 0.10714920, 0.11134535, 0…
$ q95                  <dbl> 6.373021, 7.123049, 6.854887, 8.010503, 8.095148,…
$ high_q_freq          <dbl> 6.10, 3.90, 12.25, 18.90, 14.95, 14.10, 16.05, 16…
$ high_q_dur           <dbl> 8.714286, 2.294118, 7.205882, 3.286957, 2.577586,…
$ low_q_freq           <dbl> 41.35, 65.15, 89.25, 94.80, 71.55, 58.90, 82.20, …
$ low_q_dur            <dbl> 20.170732, 17.144737, 19.402174, 14.697674, 12.77…
$ zero_q_freq          <dbl> 0.00000000, 0.00000000, 0.00000000, 0.00000000, 0…
$ hfd_mean             <dbl> 207.25, 166.25, 184.90, 181.00, 184.80, 197.20, 1…
$ soil_depth_pelletier <dbl> 7.4047619, 17.4128079, 19.0114144, 7.2525570, 5.3…
$ soil_depth_statsgo   <dbl> 1.248408, 1.491846, 1.461363, 1.279047, 1.392779,…
$ soil_porosity        <dbl> 0.4611488, 0.4159055, 0.4590910, 0.4502360, 0.422…
$ soil_conductivity    <dbl> 1.106522, 2.375005, 1.289807, 1.373292, 2.615154,…
$ max_water_content    <dbl> 0.5580548, 0.6262289, 0.6530198, 0.5591227, 0.561…
$ sand_frac            <dbl> 27.84183, 59.39016, 32.23546, 35.26903, 55.16313,…
$ silt_frac            <dbl> 55.15694, 28.08094, 51.77918, 50.84123, 34.18544,…
$ clay_frac            <dbl> 16.275732, 12.037646, 14.776824, 12.654125, 10.30…
$ water_frac           <dbl> 5.3766978, 1.2269127, 1.6343449, 0.6745936, 0.000…
$ organic_frac         <dbl> 0.4087168, 0.0000000, 1.3302776, 0.0000000, 0.000…
$ other_frac           <dbl> 0.0000000, 0.3584723, 0.0220161, 0.0000000, 0.147…
$ gauge_lat            <dbl> 47.23739, 44.60797, 45.50097, 45.17501, 44.86920,…
$ gauge_lon            <dbl> -68.58264, -67.93524, -68.30596, -69.31470, -69.9…
$ elev_mean            <dbl> 250.31, 92.68, 143.80, 247.80, 310.38, 615.70, 47…
$ slope_mean           <dbl> 21.64152, 17.79072, 12.79195, 29.56035, 49.92122,…
$ area_gages2          <dbl> 2252.70, 573.60, 3676.17, 769.05, 909.10, 383.82,…
$ area_geospa_fabric   <dbl> 2303.95, 620.38, 3676.09, 766.53, 904.94, 396.10,…
$ frac_forest          <dbl> 0.9063, 0.9232, 0.8782, 0.9548, 0.9906, 1.0000, 1…
$ lai_max              <dbl> 4.167304, 4.871392, 4.685200, 4.903259, 5.086811,…
$ lai_diff             <dbl> 3.340732, 3.746692, 3.665543, 3.990843, 4.300978,…
$ gvf_max              <dbl> 0.8045674, 0.8639358, 0.8585020, 0.8706685, 0.891…
$ gvf_diff             <dbl> 0.3716482, 0.3377125, 0.3513934, 0.3986194, 0.445…
$ dom_land_cover_frac  <dbl> 0.8834519, 0.8204934, 0.9752580, 1.0000000, 0.850…
$ dom_land_cover       <chr> "    Mixed Forests", "    Mixed Forests", "    Mi…
$ root_depth_50        <dbl> NA, 0.2374345, NA, 0.2500000, 0.2410270, 0.225615…
$ root_depth_99        <dbl> NA, 2.238444, NA, 2.400000, 2.340180, 2.237435, 2…

# Summarize the dataset
skim(camels_data)

Data summary

Name	camels_data
Number of rows	671
Number of columns	58
_______________________
Column type frequency:
character	6
numeric	52
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
gauge_id	0	1.00	8	8	0	671	0
high_prec_timing	0	1.00	3	3	0	4	0
low_prec_timing	0	1.00	3	3	0	4	0
geol_1st_class	0	1.00	12	31	0	12	0
geol_2nd_class	138	0.79	12	31	0	13	0
dom_land_cover	0	1.00	12	38	0	12	0

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
p_mean	0	1.00	3.26	1.41	0.64	2.37	3.23	3.78	8.94	▃▇▂▁▁
pet_mean	0	1.00	2.79	0.55	1.90	2.34	2.69	3.15	4.74	▇▇▅▂▁
p_seasonality	0	1.00	-0.04	0.53	-1.44	-0.26	0.08	0.22	0.92	▁▂▃▇▂
frac_snow	0	1.00	0.18	0.20	0.00	0.04	0.10	0.22	0.91	▇▂▁▁▁
aridity	0	1.00	1.06	0.62	0.22	0.70	0.86	1.27	5.21	▇▂▁▁▁
high_prec_freq	0	1.00	20.93	4.55	7.90	18.50	22.00	24.23	32.70	▂▃▇▇▁
high_prec_dur	0	1.00	1.35	0.19	1.08	1.21	1.28	1.44	2.09	▇▅▂▁▁
low_prec_freq	0	1.00	254.65	35.12	169.90	232.70	255.85	278.92	348.70	▂▅▇▅▁
low_prec_dur	0	1.00	5.95	3.20	2.79	4.24	4.95	6.70	36.51	▇▁▁▁▁
glim_1st_class_frac	0	1.00	0.79	0.20	0.30	0.61	0.83	1.00	1.00	▁▃▃▃▇
glim_2nd_class_frac	0	1.00	0.16	0.14	0.00	0.00	0.14	0.27	0.49	▇▃▃▂▁
carbonate_rocks_frac	0	1.00	0.12	0.26	0.00	0.00	0.00	0.04	1.00	▇▁▁▁▁
geol_porostiy	3	1.00	0.13	0.07	0.01	0.07	0.13	0.19	0.28	▇▆▇▇▂
geol_permeability	0	1.00	-13.89	1.18	-16.50	-14.77	-13.96	-13.00	-10.90	▂▅▇▅▂
q_mean	1	1.00	1.49	1.54	0.00	0.63	1.13	1.75	9.69	▇▁▁▁▁
runoff_ratio	1	1.00	0.39	0.23	0.00	0.24	0.35	0.51	1.36	▆▇▂▁▁
slope_fdc	1	1.00	1.24	0.51	0.00	0.90	1.28	1.63	2.50	▂▅▇▇▁
baseflow_index	0	1.00	0.49	0.16	0.01	0.40	0.50	0.60	0.98	▁▃▇▅▁
stream_elas	1	1.00	1.83	0.78	-0.64	1.32	1.70	2.23	6.24	▁▇▃▁▁
q5	1	1.00	0.17	0.27	0.00	0.01	0.08	0.22	2.42	▇▁▁▁▁
q95	1	1.00	5.06	4.94	0.00	2.07	3.77	6.29	31.82	▇▂▁▁▁
high_q_freq	1	1.00	25.74	29.07	0.00	6.41	15.10	35.79	172.80	▇▂▁▁▁
high_q_dur	1	1.00	6.91	10.07	0.00	1.82	2.85	7.55	92.56	▇▁▁▁▁
low_q_freq	1	1.00	107.62	82.24	0.00	37.44	96.00	162.14	356.80	▇▆▅▂▁
low_q_dur	1	1.00	22.28	21.66	0.00	10.00	15.52	26.91	209.88	▇▁▁▁▁
zero_q_freq	1	1.00	0.03	0.11	0.00	0.00	0.00	0.00	0.97	▇▁▁▁▁
hfd_mean	1	1.00	182.52	33.53	112.25	160.16	173.77	204.05	287.75	▂▇▃▂▁
soil_depth_pelletier	0	1.00	10.87	16.24	0.27	1.00	1.23	12.89	50.00	▇▁▁▁▁
soil_depth_statsgo	0	1.00	1.29	0.27	0.40	1.11	1.46	1.50	1.50	▁▁▂▂▇
soil_porosity	0	1.00	0.44	0.02	0.37	0.43	0.44	0.46	0.68	▃▇▁▁▁
soil_conductivity	0	1.00	1.74	1.52	0.45	0.93	1.35	1.93	13.96	▇▁▁▁▁
max_water_content	0	1.00	0.53	0.15	0.09	0.43	0.56	0.64	1.05	▁▅▇▃▁
sand_frac	0	1.00	36.47	15.63	8.18	25.44	35.27	44.46	91.98	▅▇▅▁▁
silt_frac	0	1.00	33.86	13.25	2.99	23.95	34.06	43.64	67.77	▂▆▇▆▁
clay_frac	0	1.00	19.89	9.32	1.85	14.00	18.66	25.42	50.35	▃▇▅▂▁
water_frac	0	1.00	0.10	0.94	0.00	0.00	0.00	0.00	19.35	▇▁▁▁▁
organic_frac	0	1.00	0.59	3.84	0.00	0.00	0.00	0.00	57.86	▇▁▁▁▁
other_frac	0	1.00	9.82	16.83	0.00	0.00	1.31	11.74	99.38	▇▁▁▁▁
gauge_lat	0	1.00	39.24	5.21	27.05	35.70	39.25	43.21	48.82	▂▃▇▆▅
gauge_lon	0	1.00	-95.79	16.21	-124.39	-110.41	-92.78	-81.77	-67.94	▆▃▇▇▅
elev_mean	0	1.00	759.42	786.00	10.21	249.67	462.72	928.88	3571.18	▇▂▁▁▁
slope_mean	0	1.00	46.20	47.12	0.82	7.43	28.80	73.17	255.69	▇▂▂▁▁
area_gages2	0	1.00	792.62	1701.95	4.03	122.28	329.68	794.30	25791.04	▇▁▁▁▁
area_geospa_fabric	0	1.00	808.08	1709.85	4.10	127.98	340.70	804.50	25817.78	▇▁▁▁▁
frac_forest	0	1.00	0.64	0.37	0.00	0.28	0.81	0.97	1.00	▃▁▁▂▇
lai_max	0	1.00	3.22	1.52	0.37	1.81	3.37	4.70	5.58	▅▆▃▅▇
lai_diff	0	1.00	2.45	1.33	0.15	1.20	2.34	3.76	4.83	▇▇▇▆▇
gvf_max	0	1.00	0.72	0.17	0.18	0.61	0.78	0.86	0.92	▁▁▂▃▇
gvf_diff	0	1.00	0.32	0.15	0.03	0.19	0.32	0.46	0.65	▃▇▅▇▁
dom_land_cover_frac	0	1.00	0.81	0.18	0.31	0.65	0.86	1.00	1.00	▁▂▃▃▇
root_depth_50	24	0.96	0.18	0.03	0.12	0.17	0.18	0.19	0.25	▃▃▇▂▂
root_depth_99	24	0.96	1.83	0.30	1.50	1.52	1.80	2.00	3.10	▇▃▂▁▁

# Visualize missing data
vis_miss(camels_data)

# Drop variables with more than 20% missing data
camels_clean <- camels_data %>%
  select(where(~ mean(!is.na(.)) > 0.8)) %>%
  drop_na(q_mean)  # Ensure response variable is present

# Look at correlation of numeric variables
numeric_vars <- camels_clean %>% select(where(is.numeric))

# Plot correlation matrix
cor_matrix <- cor(numeric_vars, use = "complete.obs")
ggcorrplot(cor_matrix, lab = TRUE, lab_size = 2.5)

# Set a seed for reproducibility
set.seed(123)

# Split the cleaned dataset: 80% training, 20% testing
camels_split <- initial_split(camels_clean, prop = 0.8)

# Extract training and testing sets
camels_train <- training(camels_split)
camels_test <- testing(camels_split)

# Optional: check dimensions
dim(camels_train)

[1] 536  57

dim(camels_test)

[1] 134  57

library(recipes)

recipe_data <- recipe(~ ., data = camels_data) %>%
  step_zv(all_predictors()) %>%  # Remove zero variance predictors
  step_novel(all_nominal()) %>%  # Handle new levels in categorical variables
  step_unknown(all_nominal()) %>%  # Handle unknown levels in categorical variables
  step_corr(all_predictors(), threshold = 0.9) %>%  # Remove highly correlated predictors
  step_impute_median(all_numeric()) %>%  # Impute missing numeric values using median
  step_impute_mode(all_nominal()) %>%  # Impute missing nominal values using mode
  step_dummy(all_nominal(), -all_outcomes()) %>%  # Create dummy variables for nominal features
  step_normalize(all_numeric())  # Normalize numeric predictors

library(tidymodels)

# Clean the training data by removing rows with missing or infinite q_mean
camels_train_cleaned <- camels_train %>%
  filter(!is.na(q_mean) & !is.infinite(q_mean) & q_mean != -Inf & q_mean != Inf) %>%
  drop_na()  # Drop any other rows with missing values

library(recipes)

camels_recipe <- recipe(runoff_ratio ~ ., data = camels_train) %>%
  step_zv() %>%  # Remove predictors with zero variance
  step_novel(all_nominal_predictors()) %>%
  step_unknown(all_nominal_predictors()) %>%
  step_dummy(all_nominal_predictors()) %>%
  step_normalize(all_numeric_predictors())

# Create 10-fold cross-validation folds
camels_folds <- vfold_cv(camels_train_cleaned, v = 10)

# Define model specifications without tuning
lin_reg_spec <- linear_reg() %>%
  set_engine("lm") %>%
  set_mode("regression")

# Random Forest with fixed parameters
rf_spec <- rand_forest(mtry = 3, trees = 500) %>%
  set_engine("ranger") %>%
  set_mode("regression")

# XGBoost with fixed parameters
xgb_spec <- boost_tree(mtry = 3, trees = 500, learn_rate = 0.1) %>%
  set_engine("xgboost") %>%
  set_mode("regression")

# Create the workflow set
workflow_set <- workflow_set(
  preproc = list(camels = camels_recipe),
  models = list(
    lin_reg = lin_reg_spec,
    random_forest = rf_spec,
    xgboost = xgb_spec
  )
)

# Fit the workflows with cross-validation and evaluate using RMSE and R-squared
results <- workflow_map(
  workflow_set,
  resamples = camels_folds,
  metrics = metric_set(rmse, rsq),
  verbose = TRUE
)

i   No tuning parameters. `fit_resamples()` will be attempted

i 1 of 3 resampling: camels_lin_reg

→ A | warning: !  The following columns have zero variance so scaling cannot be used:
                 gauge_id_new, gauge_id_unknown, high_prec_timing_new,
                 high_prec_timing_unknown, low_prec_timing_new, low_prec_timing_unknown,
                 geol_1st_class_new, geol_1st_class_unknown, dom_land_cover_new, and
                 dom_land_cover_unknown.
               ℹ Consider using ?step_zv (`?recipes::step_zv()`) to remove those columns
                 before normalizing.

There were issues with some computations   A: x1

→ B | warning: prediction from rank-deficient fit; consider predict(., rankdeficient="NA")

There were issues with some computations   A: x1
There were issues with some computations   A: x1   B: x1
There were issues with some computations   A: x10   B: x10

✔ 1 of 3 resampling: camels_lin_reg (2.8s)
i   No tuning parameters. `fit_resamples()` will be attempted
i 2 of 3 resampling: camels_random_forest
→ A | warning: !  The following columns have zero variance so scaling cannot be used:
                 gauge_id_new, gauge_id_unknown, high_prec_timing_new,
                 high_prec_timing_unknown, low_prec_timing_new, low_prec_timing_unknown,
                 geol_1st_class_new, geol_1st_class_unknown, dom_land_cover_new, and
                 dom_land_cover_unknown.
               ℹ Consider using ?step_zv (`?recipes::step_zv()`) to remove those columns
                 before normalizing.
There were issues with some computations   A: x3
There were issues with some computations   A: x10

✔ 2 of 3 resampling: camels_random_forest (2.1s)
i   No tuning parameters. `fit_resamples()` will be attempted
i 3 of 3 resampling: camels_xgboost
→ A | warning: !  The following columns have zero variance so scaling cannot be used:
                 gauge_id_new, gauge_id_unknown, high_prec_timing_new,
                 high_prec_timing_unknown, low_prec_timing_new, low_prec_timing_unknown,
                 geol_1st_class_new, geol_1st_class_unknown, dom_land_cover_new, and
                 dom_land_cover_unknown.
               ℹ Consider using ?step_zv (`?recipes::step_zv()`) to remove those columns
                 before normalizing.
There were issues with some computations   A: x4
There were issues with some computations   A: x10

✔ 3 of 3 resampling: camels_xgboost (3.8s)

# Visualize resampling results
autoplot(results)

Model Selection Based on the visualized metrics, select a model that you think best performs. Describe the reason for your choice using the metrics.

I chose Boosted Trees because it has the lowest errors with cross-validation and has the highest R-squared.

Describe the model you selected. What is the model type, engine, and mode. Why do you think it is performing well for this problem?

Model Type: Boosted Trees Engine: xgboost Mode: regression It is performing well because XGBoost is good for non-linear relationships, interactions between predictors, dealing with missing values, and avoiding over fitting.

library(tidymodels)

# Tunable XGBoost model specification
xgb_tune_spec <- boost_tree(
  trees = 500,                        # fixed number of trees
  mtry = tune(),                      # tunable: number of variables randomly selected
  learn_rate = tune(),               # tunable: step size shrinkage
  tree_depth = tune()                # tunable: max depth of trees
) %>%
  set_engine("xgboost") %>%
  set_mode("regression")

library(workflows)

# Create a workflow object combining the recipe and the tunable XGBoost model
xgb_workflow <- workflow() %>%
  add_model(xgb_tune_spec) %>%
  add_recipe(camels_recipe)

# Load required library
library(tune)

# Extract tunable parameters and their ranges
dials <- extract_parameter_set_dials(xgb_workflow)

# View the parameter objects and their ranges
dials$object

[[1]]

# Randomly Selected Predictors (quantitative)

Range: [1, ?]

[[2]]

Tree Depth (quantitative)

Range: [1, 15]

[[3]]

Learning Rate (quantitative)

Transformer: log-10 [1e-100, Inf]

Range (transformed scale): [-3, -0.5]

# Load required libraries
library(tidymodels)
library(dials)

tune_spec <- rand_forest(
  mtry = tune(),         # Number of predictors to sample
  min_n = tune(),        # Minimum number of observations in a node
  trees = 500            # Fixed number of trees
) %>%
  set_engine("ranger") %>%
  set_mode("regression")

wf_tune <- workflow() %>%
  add_recipe(camels_recipe) %>%
  add_model(tune_spec)

# Ensure the k-fold cross-validation folds are defined (e.g., camels_folds)
camels_folds <- vfold_cv(camels_train_cleaned, v = 5)  # Example with 5-fold cross-validation

# Extract tunable parameters from the workflow
dials <- extract_parameter_set_dials(wf_tune)

# Optional: If you're using mtry (which depends on number of predictors), finalize it:
dials <- finalize(dials, camels_train_cleaned)

# Now create the grid
my.grid <- grid_space_filling(
  dials,   # No need for $parameters
  size = 25
)

# Tune the model using grid search
model_params <- tune_grid(
  wf_tune,
  resamples = camels_folds,  # Ensure camels_folds is defined as your k-folds
  grid = my.grid,  # Grid of hyperparameters
  metrics = metric_set(rmse, rsq, mae),  # Define evaluation metrics
  control = control_grid(save_pred = TRUE)  # Save predictions
)

→ A | warning: !  The following columns have zero variance so scaling cannot be used:
                 gauge_id_new, gauge_id_unknown, high_prec_timing_new,
                 high_prec_timing_unknown, low_prec_timing_new, low_prec_timing_unknown,
                 geol_1st_class_new, geol_1st_class_unknown, dom_land_cover_new, and
                 dom_land_cover_unknown.
               ℹ Consider using ?step_zv (`?recipes::step_zv()`) to remove those columns
                 before normalizing.

There were issues with some computations   A: x1

There were issues with some computations   A: x2

There were issues with some computations   A: x3

There were issues with some computations   A: x4

There were issues with some computations   A: x5
There were issues with some computations   A: x5

# Visualize the tuning results
autoplot(model_params)

collect_metrics(model_params)

# A tibble: 75 × 8
    mtry min_n .metric .estimator   mean     n std_err .config              
   <int> <int> <chr>   <chr>       <dbl> <int>   <dbl> <chr>                
 1     1    24 mae     standard   0.165      5 0.00943 Preprocessor1_Model01
 2     1    24 rmse    standard   0.217      5 0.0128  Preprocessor1_Model01
 3     1    24 rsq     standard   0.719      5 0.0241  Preprocessor1_Model01
 4     3    13 mae     standard   0.124      5 0.00863 Preprocessor1_Model02
 5     3    13 rmse    standard   0.170      5 0.0126  Preprocessor1_Model02
 6     3    13 rsq     standard   0.757      5 0.0343  Preprocessor1_Model02
 7     5    33 mae     standard   0.100      5 0.00685 Preprocessor1_Model03
 8     5    33 rmse    standard   0.141      5 0.0111  Preprocessor1_Model03
 9     5    33 rsq     standard   0.787      5 0.0284  Preprocessor1_Model03
10     8     5 mae     standard   0.0792     5 0.00557 Preprocessor1_Model04
# ℹ 65 more rows

library(dplyr)

# View metrics ordered by lowest MAE (Mean Absolute Error)
collect_metrics(model_params) %>%
  filter(.metric == "mae") %>%
  arrange(mean)

# A tibble: 25 × 8
    mtry min_n .metric .estimator   mean     n std_err .config              
   <int> <int> <chr>   <chr>       <dbl> <int>   <dbl> <chr>                
 1    50     6 mae     standard   0.0482     5 0.00524 Preprocessor1_Model22
 2    54    14 mae     standard   0.0490     5 0.00532 Preprocessor1_Model24
 3    38     2 mae     standard   0.0507     5 0.00529 Preprocessor1_Model17
 4    52    22 mae     standard   0.0509     5 0.00509 Preprocessor1_Model23
 5    40     9 mae     standard   0.0512     5 0.00527 Preprocessor1_Model18
 6    57    32 mae     standard   0.0515     5 0.00524 Preprocessor1_Model25
 7    43    17 mae     standard   0.0523     5 0.00507 Preprocessor1_Model19
 8    45    28 mae     standard   0.0539     5 0.00534 Preprocessor1_Model20
 9    29     8 mae     standard   0.0549     5 0.00541 Preprocessor1_Model13
10    47    38 mae     standard   0.0557     5 0.00535 Preprocessor1_Model21
# ℹ 15 more rows

show_best(model_params, metric = "mae")

# A tibble: 5 × 8
   mtry min_n .metric .estimator   mean     n std_err .config              
  <int> <int> <chr>   <chr>       <dbl> <int>   <dbl> <chr>                
1    50     6 mae     standard   0.0482     5 0.00524 Preprocessor1_Model22
2    54    14 mae     standard   0.0490     5 0.00532 Preprocessor1_Model24
3    38     2 mae     standard   0.0507     5 0.00529 Preprocessor1_Model17
4    52    22 mae     standard   0.0509     5 0.00509 Preprocessor1_Model23
5    40     9 mae     standard   0.0512     5 0.00527 Preprocessor1_Model18

hp_best <- select_best(model_params, metric = "mae")

final_wf <- finalize_workflow(
  wf_tune,
  hp_best
)

final_fit <- last_fit(
  final_wf,        # your finalized workflow
  split = camels_split  # original data split from initial_split()
)

→ A | warning: !  The following columns have zero variance so scaling cannot be used:
                 gauge_id_new, gauge_id_unknown, high_prec_timing_new,
                 high_prec_timing_unknown, low_prec_timing_new, low_prec_timing_unknown,
                 geol_1st_class_new, geol_1st_class_unknown, dom_land_cover_new, and
                 dom_land_cover_unknown.
               ℹ Consider using ?step_zv (`?recipes::step_zv()`) to remove those columns
                 before normalizing.

metrics_test <- collect_metrics(final_fit)
print(metrics_test)

# A tibble: 2 × 4
  .metric .estimator .estimate .config             
  <chr>   <chr>          <dbl> <chr>               
1 rmse    standard      0.0622 Preprocessor1_Model1
2 rsq     standard      0.921  Preprocessor1_Model1

predictions_test <- collect_predictions(final_fit)

library(ggplot2)

# Create scatter plot with actual vs predicted values
ggplot(predictions_test, aes(x = runoff_ratio, y = .pred)) +
  geom_point(aes(color = runoff_ratio), alpha = 0.6) +
  geom_smooth(method = "lm", se = FALSE, color = "blue") +
  geom_abline(intercept = 0, slope = 1, color = "red") +
  scale_color_viridis_c() +
  labs(
    title = "Predicted vs Actual Values",
    x = "Actual Values (Runoff Ratio)",
    y = "Predicted Values"
  ) +
  theme_minimal()

`geom_smooth()` using formula = 'y ~ x'

camels_data_clean <- camels_data %>%
  filter(!is.na(runoff_ratio))

# Fit the finalized workflow to the full cleaned data
final_fit_full <- fit(final_wf, data = camels_data_clean)

Warning: !  The following columns have zero variance so scaling cannot be used:
  gauge_id_new, gauge_id_unknown, high_prec_timing_new,
  high_prec_timing_unknown, low_prec_timing_new, low_prec_timing_unknown,
  geol_1st_class_new, geol_1st_class_unknown, dom_land_cover_new, and
  dom_land_cover_unknown.
ℹ Consider using ?step_zv (`?recipes::step_zv()`) to remove those columns
  before normalizing.

library(broom)
augmented_preds <- augment(final_fit_full, new_data = camels_data_clean)

library(dplyr)
augmented_preds <- augmented_preds %>%
  mutate(residual = (.pred - runoff_ratio)^2)  # or .pred - runoff_ratio for raw residuals

library(ggplot2)
library(patchwork)

# Prediction map
pred_map <- ggplot(augmented_preds, aes(x = gauge_lon, y = gauge_lat, color = .pred)) +
  geom_point(size = 2) +
  scale_color_viridis_c(option = "C") +
  labs(title = "Predicted Runoff Ratio", color = "Prediction") +
  theme_minimal()

# Residual map
resid_map <- ggplot(augmented_preds, aes(x = gauge_lon, y = gauge_lat, color = residual)) +
  geom_point(size = 2) +
  scale_color_viridis_c(option = "A") +
  labs(title = "Residuals (Squared)", color = "Residual") +
  theme_minimal()

# Combine with patchwork
pred_map + resid_map